Clustering Analysis of School Grant Recipients Based on Grant Allocation Characteristics Using the K-Means Algorithm in North Sumatra Province

Abstract

The distribution of educational grants requires systematic analysis to understand funding
allocation patterns among educational institutions. This study aims to analyze the
characteristics of school grant recipients in North Sumatra Province using the K-Means
clustering algorithm. The study uses administrative grant data covering the 2023–2026 period,
consisting of recipient institution names and addresses and nominal grant amounts. The
research process includes data preprocessing, numerical feature selection, data transformation,
K-Means clustering, and cluster evaluation. The grant amounts are transformed using the
natural logarithm of one plus the nominal value to reduce the influence of extreme values in
the clustering process. The algorithm is configured to generate three clusters representing
lower, moderate, and higher grant allocation levels. The clustering results are evaluated using
the Silhouette Coefficient to provide an initial assessment of cluster cohesion and separation.
The findings are expected to provide a structured overview of grant allocation characteristics
and support the descriptive evaluation of educational funding distribution. However, the
resulting clusters represent differences in nominal grant amounts rather than direct
measurements of institutional needs or funding priorities. Further research incorporating
student enrollment, infrastructure conditions, socioeconomic characteristics, and educational
facility availability is necessary to support a more comprehensive, evidence-based approach
to grant allocation.